• Announcements

• Current issues:
• VSR4 readiness:
  
• code freeze, requests, ...
• data transfer preparation (Livio)
• RDS production (Fscan, STAMP). Production and transfer before the end of the run?
• LSC-Virgo low latency transfer: update (Benoit, ....)
  
• Data transfer/data access at CNAF and CCIN2P3
• update on VSR2/VSR3 pre-staging at CNAF (Alberto, ...)
  
• Lyon data access (MAB, ...)

• Specific topics:
  
• CMT/autoconf interoperability:
• CMT future? (MAB): CMake, waf
  
• Adv Detector software developments:
• definition of a computing model
•  workflow
• Data Analysis white paper: computing & software

Minutes:

1. VSR4 preparation:

Franco: Fr/v8r11 has been kept. v8r14 switch would have been too much work. Only Dy is using v8r14. Most of the applications have been restarted from /virgoApp. What's remaining? online processing (hrec applications) are still missing from virgoApp. Started from virgoDev.
Benoit: Frv and Fd needs to be archived. Will do that before the end of the day. Then you can restart applications from /virgoApp.
Franco: MBTA during the run?
Benoit: just restarted now. But no time to move it in /virgoApp. will start from /virgoDev. Will need some weeks to archive in CVS. but MBTA will be only useful for detchar purposes. BTW, Fr/v8r14 affects only offline. The DAQ does not need this version.

Data transfer:
Antonella: at CNAF we need some space. --> MAB will contact Alberto
At Lyon: 1 month of delay that could be recovered after the run. No data loss.
Decision: on friday restart real time transfer.

Data Access at CCs:
- No news from CNAF ... good news!
Franco: data sender running at CNAF. Tests OK. Need now feedback from users.

- Lyon test: MAB tried to use fopen() instead of open () but this is not compatible with XrootD which is working only with POSIX functions.
Benoit: did you try v8r13? could help timeout issues ...
MAB: yes. Same pb. I'm now in contact with Lyon experts.

RDS production:
Benoit : online production request is coming too late. Will start to produce RDS in the coming weeks but offline.
Stefano: data size ?
Benoit : rather small. Few % of rawdata size. Will use a procdata disk.


Low latency transfer meeting:
Benoit: the committee started to collect information about requirements/constraints (astro, DAQ,etc)
very low latency transfer (done by Fd during S6/VSR3). Will start discussion about technical soluations in the next call. Everyone is invited to attend these meetings. Recall: during S6/VSR3 we had an offset of 40s due mainly to h(t) production. People are now talking about 10s of delay. Going below does not look reasonable technically.

MAB: about the technical proposals, are there any issues about security (certificates, ...) required for new generation of low latency data transfer? Data could be easily sniffed out on the network.
Benoit: this has not been addressed. Robustness is more a concern. As far as security is concerned, it's true that someone expert, knowing the inside information (ports used/machines, etc ...) could steal the data, but this must come from inside the collaboration ...

Adjourn the meeting here. AdvVirgo computing plan requires DA white paper written.